LA COMUNICACIÓN INTERPERSONAL

The clearance time varies between cell types in vitro and probably in vivo,

and microglia are the most proficient phagocytes

The digestion phase o f clearance in vitro showed generally less variation between cell types than the recognition phase. Microglia were the most proficient phagocytes, digesting all engulfed cells in under 2 hours (fig. 3.3b). LECs and BHKs were slower, but still digested the majority o f engulfed cells to completion within 4 hours. Astrocytes however, were extremely poor at clearance in culture and most o f the cells engulfed by them had not been cleared by the end o f the experiments (> 24 hours). This could reflect a real inefficiency in the ability o f astrocytes to clear cells, or result from the conditions o f the culture system, and the same could apply to the LECs and BHKs.

However, differences in the ability to degrade cells were also inferred from the electron micrographs o f pyknoses in the optic nerve and the EGL. Pyknoses in the late, floculent stages o f pyknosis formed a 5-fold greater proportion o f the total in the EGL than in the optic nerve, implying that their final dissolution by Bergmann gha and bystander neuroblasts is a more protracted event process than in microglia. The phagocytic granule neurons consist o f little more than a nucleus with a small amount o f ribosome packed cytoplasm (Peters et a i, 1991), and it is reasonable to think that the digestion o f a large pyknotic cell must represent a great burden on their lytic machinery and that it would progress more slowly than in a ly so some-packed microglia.

Different cells bave different clearance times: consequences in vivo

Summing the recognition and digestion phases (fig. 3.3c) gave the total clearance time and this chapter presents the first direct evidence that different cells have different abilities to clear pyknoses, at least in vitro. Indirect evidence from

comparing pyknotic figures from the EGL and optic nerve suggests that clearance occurs at different rates in vivo as well as in vitro. The implication o f this is that one cannot evaluate the amount o f cell death in tissue from its pyknotic index alone. Both the cell types involved in clearance, and the fraction o f the total pyknoses found in each cell type must be known. Returning to the situation in the P7 rat cerebellar white matter (chapter 2), where the number o f pyknoses at P i 0 is half what it was at P7, one can see that it is possible that the amount o f cell death may not halve between P7 and PIO, as the total number o f pyknoses suggests, because the fraction o f pyknoses inside microglia increases 3-fold and they may actually be several times as efficient at clearance than astrocytes. In fact, if one were to use the in vitro data, where microglia are 10-fold more efficient at clearance than astrocytes, the amount o f cell death in the cerebellar white matter would actually be increasing over this period (calculating the ratio o f PCD at P10:P7 given that the number o f pyknoses falls by half, that microglia phagocytose 3/4 at PIG and 1/4 at P7, and that microglia take 10-fold less time to clear a pyknosis than an astrocyte results in the amount o f cell death increasing by approximately 20% over this period).

A second important implication o f the results o f this chapter is that pyknotic indices o f over 2% need to be examined carefiilly because they could be due to 2 very different reasons:

i) The amount o f cell death really is substantial and that clearance is taking place rapidly, probably by cells o f the macrophage/monocyte lineage. If cells are

cleared in one hour, a pyknotic index o f 2% implies that 50% o f the tissue is cleared daily. This kind o f huge level o f cell death can take place only where the rate o f production o f cells is equally rapid, such as in the neonatal thymus, where the pyknotic index can be as high as 2% (Surh and Sprent, 1994), or when a tissue

is undergoing involution, which is what happens during interdigital regression. An important figure to remember is that, if cells are being cleared in one hour, a pyknotic index o f only 4% results in 100% o f the cells in that tissue being cleared everyday.

ii) Pyknotic indices above a fraction o f a percent thus imply large scale cell death or a relatively slow clearance time. The P7 cerebellar white matter has a

pyknotic index o f 2% (Krueger et al., 1995). If the clearance time was 1 hour, as the authors speculate, this represents a turnover o f 50% o f the white matter every day. Since the tissue does not regress, these cells must be made up for by cell division implying that every remaining cell has to divide every day merely to keep the tissue the same size. An alternative explanation is simply that cell death is proceeding in the white matter at a much more sedate pace and that clearance times are generally much longer than one hour. Kreuger et al. used an estimate o f clearance time based on an extrapolation o f the clearance time in the optic nerve (Barres et al, 1992). We now know that the exclusive phagocytes o f the optic nerve are microglia, which are particularly efficient at clearing pyknoses and that it is probably unreasonable to assume that clearance happens at the same rate in other tissues, where microglia and macrophages operate in concert with other phagocytes. In the white matter, 3/4 o f the pyknoses at P7 are inside astrocytes. If these were to have a clearance time about 10-fold longer than the microglia, their contribution could essentially be ignored in the first instance resulting in a pyknotic index o f 0.5% inside the microglia. This would now give a rate o f cell death o f 10% o f the cells in the white matter per day. This is still a high level of death and tissue sculpting, but it abrogates the need for every cell to divide every day to maintain the size o f the tissue.

Chapter 4: